基于单应性校正P-KLT的无人机视频稳像
|
摘要
手持设备和无人机拍摄的视频存在不可避免的抖动,降低视频观看及后期处理的效果。视频稳像技术中考虑到实时性,常用相似矩阵变换来进行帧间校正,对于三维运动的补偿效果较差,但是直接采用可以表示三维运动关系的单应性矩阵变换,实时性无法满足要求。对此,提出了一种基于单应性校正P-KLT的视频稳像算法,此算法是以光流跟踪理论为基础,对单应性矩阵变换实现实时参数跟踪计算,针对单应性变换存在图像局部畸变的问题,提出了基于顶点滤波的畸变校正算法,满足了三维运动视频实时稳像要求。实验结果表明,经过本算法处理后的视频画面稳定自然,尤其是对存在运动模糊的视频也有较好的处理结果。
There are inevitable shakes in videos taken by handheld devices and UAVs, which reduce the effect of video viewing and post-processing. In the video stabilization technology, similar matrix transformation is often used to perform inter-frame correction to have a real time processing. But the compensation effect for three-dimensional motion is poor to use similar matrix transformation. On the other hand, the homography matrix transformation that can represent three-dimensional motion, can not meet the real time requirements. To solve this problem, a video stabilization technique based on homography revised P-KLT is proposed. This algorithm is based on the optical flow tracking theory and realizes real-time parameter tracking calculation for the homography matrix transformation. A distortion correction algorithm based on vertex filtering is proposed to meet the requirements of 3D motion video real-time image stabilization. The experimental results show that the video images processed by this algorithm are stable and natural, especially for videos with motion blur.
There are inevitable shakes in videos taken by handheld devices and UAVs, which reduce the effect of video viewing and post-processing. In the video stabilization technology, similar matrix transformation is often used to perform inter-frame correction to have a real time processing. But the compensation effect for three-dimensional motion is poor to use similar matrix transformation. On the other hand, the homography matrix transformation that can represent three-dimensional motion, can not meet the real time requirements. To solve this problem, a video stabilization technique based on homography revised P-KLT is proposed. This algorithm is based on the optical flow tracking theory and realizes real-time parameter tracking calculation for the homography matrix transformation. A distortion correction algorithm based on vertex filtering is proposed to meet the requirements of 3D motion video real-time image stabilization. The experimental results show that the video images processed by this algorithm are stable and natural, especially for videos with motion blur.
引文
[1]Wang Y,Hou Z J,Leman.Real-Time Video Stabilization for Unmanned Aerial Vehicles[C]//Machine Vision Applications,2011,IAPR Conference on.IAPR,2011:336-339.
[2]王志民,徐晓刚.电子稳像技术综述.中国图象图形学报,2010,15(3):470-480.Wang Z M,Xu X G.A Survey on Electronic Image Stabilization[J].Journal of Image and Graphics,2010,15(3):470-480.
[3]蔡成涛,梁小龙,谭吉来.自适应最优块匹配视频稳像算法.系统工程与电子技术,2013,35(6):1 324-1 329.Cai C T,Liang X L,Tan J L.Adaptive Optimal Block Matching Video Stabilization Algorithm[J].System Engineering and Electronics,2013,35(6):1324-1329
[4]Grundmann M,Kwatra V,Castro D,et al.Calibration-free Rolling Shutter Removal[C]//Computational Photography,2012,IEEE International Conference on.IEEE,2012:1-8.
[5]Harris C,Stephens M.A Combined Corner and Edge Detector[C]//Alvey vision,1988,Alvey vision Conference on.Alvey vision,147-151.
[6]Liu F,Jin H L,Gleicher M,et al.Content-preserving Warps for 3d Video Stabilization[J].ACM Transactions on Graphics.2009.28(3):1-9.
[7]Lucas B,Kanade T.An Iterative Image Registration Technique with an Application to Stereo Vision[J].Proceedings of Imaging Understanding Workshop,1981,81(7):674-679.
[8]Bouguet J Y.Pyramidal Implementation of the Affine Lucas Kanade Feature Tracker Description of the Algorithm[R].California:Intel Corporation,2001.
[9]Fischler M A,Bolles R C.Random Sample Consensus:a Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography[J].Communications of the ACM,1981,24(6):381-395.
[10]Kalman R E,Rudolph E.A New Approach to Linear Filtering and Prediction Problems[J].Journal of basic Engineering,1960,82(1):35-45.
[11]Grudmann M,Kwatra V,ESSA I.Auto-directed Video Stabilization with Robust L1 Optimal Camera Paths[C]//Computer Vision and Pattern Recognition,2011.CVPR 2011.IEEE Conference on.IEEE,2011.p.225-232.
[2]王志民,徐晓刚.电子稳像技术综述.中国图象图形学报,2010,15(3):470-480.Wang Z M,Xu X G.A Survey on Electronic Image Stabilization[J].Journal of Image and Graphics,2010,15(3):470-480.
[3]蔡成涛,梁小龙,谭吉来.自适应最优块匹配视频稳像算法.系统工程与电子技术,2013,35(6):1 324-1 329.Cai C T,Liang X L,Tan J L.Adaptive Optimal Block Matching Video Stabilization Algorithm[J].System Engineering and Electronics,2013,35(6):1324-1329
[4]Grundmann M,Kwatra V,Castro D,et al.Calibration-free Rolling Shutter Removal[C]//Computational Photography,2012,IEEE International Conference on.IEEE,2012:1-8.
[5]Harris C,Stephens M.A Combined Corner and Edge Detector[C]//Alvey vision,1988,Alvey vision Conference on.Alvey vision,147-151.
[6]Liu F,Jin H L,Gleicher M,et al.Content-preserving Warps for 3d Video Stabilization[J].ACM Transactions on Graphics.2009.28(3):1-9.
[7]Lucas B,Kanade T.An Iterative Image Registration Technique with an Application to Stereo Vision[J].Proceedings of Imaging Understanding Workshop,1981,81(7):674-679.
[8]Bouguet J Y.Pyramidal Implementation of the Affine Lucas Kanade Feature Tracker Description of the Algorithm[R].California:Intel Corporation,2001.
[9]Fischler M A,Bolles R C.Random Sample Consensus:a Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography[J].Communications of the ACM,1981,24(6):381-395.
[10]Kalman R E,Rudolph E.A New Approach to Linear Filtering and Prediction Problems[J].Journal of basic Engineering,1960,82(1):35-45.
[11]Grudmann M,Kwatra V,ESSA I.Auto-directed Video Stabilization with Robust L1 Optimal Camera Paths[C]//Computer Vision and Pattern Recognition,2011.CVPR 2011.IEEE Conference on.IEEE,2011.p.225-232.